The Term Premium as a Leading Macroeconomic Indicator

Forecasting the evolution path of macroeconomic variables has always been of keen interest to policy authorities. A common tool in the relevant forecasting literature is the term spread of Treasury bond interest rates. In this paper we decompose the term spread of treasury bonds into an expectations and a term premium component and we evaluate the informational content of each component in forecasting the real GDP growth rate and inflation (as measured by the GDP deflator) in various forecasting horizons. In doing so, we evaluate alternative decomposition procedures, introduce the nonlinear machine learning Support Vector Regression (SVR) methodology in rolling regressions and examine both point and conditional probability distribution forecasts. We also consider a number of control variables that are typically used in this context. According to our empirical findings neither the term spread nor its decomposition possess the ability to forecast output growth or inflation.